Inertial Measurement Unit (IMU) is used in devices to measure velocity, orientation and gravitational force. Earlier, IMU consists of two sensors namely accelerometers and gyroscopes. The accelerometer is used to measure the inertial acceleration whereas gyroscope measures angular rotation. These sensors have three degrees of freedom to measure from three axes.
Later, IMU advanced with magnetometer which measures the magnetic direction and helps to improve reading of gyroscope. IMU applications include in various areas such as tracking, navigation and robotics. IMU helps us to determine instantaneous position, velocity, orientation, and direction of movement of an object or vehicle.
Types of IMU
IMU with Two Sensor — This type of IMU consists of accelerometer and gyroscope. Each sensor has two or three degrees of freedom (DOF) defined for x, y and z-axis, which on combining for both sensors adds up to four or six degrees of freedom. The value of acceleration measured from the accelerometer and angular velocity measured from gyroscope are noted separately, while the angle measured using both sensors are calibrated to get accurate data. IMU of this kind is not interfered by external magnetic field if present around the sensor. Its accuracy is affected due to sensors noise and drift issue of the gyroscope.
IMU with Three Sensor — This type of IMU consists of accelerometer, gyroscope and magnetometer. All the sensor has 3 DOF in correspondence to 3 different axes summing up to 9 DOF in total. The magnetometer measures yaw angle rotation (i.e. rotation around the vertical axis) which is calibrated to gyroscope data to solve drift issue. Mainly use for dynamic orientation calculation when drift error is less. If IMU is present inside the magnetic field then its measurement might be affected as it uses the magnetometer.
IMU measures two to six DOF (Degrees of freedom represent a direction in which independent motion can occur). The maximum DOF it can measure is 6 which include 3 degrees of translation motion over 3 perpendicular axes and 3 degrees of rotational movement about 3 perpendicular axes (roll, pitch, and yaw). This six independent data together define the movement of an object or vehicle.
Roll, Pitch and Yaw — also known as Axes of Rotation or Principal Axes .
- Roll: also known as longitudinal axis, is the rotation of an object or vehicle on the front to back axis.
- Pitch: also called the lateral or transverse axis, is the rotation of an object or vehicle fixed between side to side (right-left) axes.
- Yaw: is the rotation around the vertical axis, lies perpendicular to an object.
An accelerometer is of various type, generally used in IMU technologies are mechanical accelerometer and piezo-electric accelerometer. A mechanical accelerometer consists of a mass suspended by springs. The displacement of the mass is measured which provides a signal that is proportional to the force F acting on the mass in the direction of the input axis. Using Newton’s second law F = ma the acceleration acting on the device is calculated.
Whereas Piezo-electric accelerometer works on the principle of piezoelectric effect, the ability of materials to get electrically polarised when subjected to mechanical stress. The PE accelerometer uses a PE element attached with a loading mass (seismic mass) to form a 1 DOF mass-spring system. The system is made one for each direction (left-right, forwards-backwards and up-down in a three-axis accelerometer). An instantaneous change in stress on the PE element produces a charge at the accelerometer’s output terminals proportional to the applied acceleration.
Similarly, gyros are of different types, but in IMU MEMS-Based Gyroscopes (MEMS) are implemented. MEMS (Micro-Electro-Mechanical Systems) is a technology defined as miniaturized mechanical and electromechanical devices and structure made up of microfabrication techniques, its dimension varies from below one micron to several millimetres.
The gyroscope uses Coriolis Effect, which states that in a frame of reference rotating at angular velocity w, a mass m moving with velocity v experiences a force, Fc = −2m(w× v). MEMS gyroscopes contain vibrating elements (such as vibrating wheel and tuning fork gyroscopes) to measure the Coriolis Effect. The simplest geometry consists of a single mass driven with a particular velocity to vibrate along a drive axis, and when an external angular rotation is applied (secondary rotation), then due to Coriolis force a secondary vibration is induced along the perpendicular axis. The angular velocity can be calculated by measuring this secondary rotation.
About 90% of the magnetometer works on the Hall Effect.
The Hall Effect principle states that when a current-carrying conductor is placed in a magnetic field, a voltage will be generated perpendicular to the direction of the field and the flow of current.
When a constant current is passed through a sheet of semiconducting material, no potential difference at the output terminal in the absence of a magnetic field. However, when a perpendicular magnetic field is present, the direction of current flow is disturbed which in turn creates a potential difference across the output terminals. This voltage is known as the Hall voltage. Keeping the input current constant the strength of the magnetic field is measured which is in proportion to Hall voltage.
Bit of Calibration
IMUs combine input data from several different sensors to accurately measure the movement. To get the accurate values calibration of the sensor is essential instead of using the raw data. Calibration parameters can be stored in the memory of an IMU and are automatically reflected in the resulted data. Calibrations are also accomplished by using a magnetometer which reduces orientation drift (errors which accumulate over time). Some device uses a proprietary sensor fusion algorithm to combine magnetometer and gyroscope data, determining the orientation of the device relative to a global frame of reference. Sensor fusion refers to processes in which signals from two or more types of sensor are used to update or maintain the state (orientation, velocity and displacement) of a system. A sensor fusion algorithm maintains this state using IMU accelerometer and gyroscope signals together with signals from additional sensors (like magnetometer) or sensor systems. The most popular techniques for performing sensor fusion is Kalman filters.
Kalman filter is an optimization algorithm to estimate the state of a system with noise and uncertainties. This filter receives unprecise measures with noise, it is able to estimate the current state with good precision and even predict the future state. The Kalman filter estimates orientation angles using all of the sensor axis contributions within the IMU and thus minimises the drift.
IMUs are used in motion sensing, unmanned navigation systems, vibration control, surveying, and various tracking systems. In unmanned navigation system (drone, aircraft), IMU is implemented to calculate altitude and relative position to a reference frame. In the manufacturing industry, IMU has been made to improve productivity. In the medical area, many IMUs has been currently used to help and assist human in surgical, or medical analysis. In robotics, some of the robots require linear and angular data for their movements; therefore the IMU is used to obtain the desired data.
The commonly known device used in the navigation system, the Global Positioning System (GPS) is very widely used in car and aircraft navigation to go from one location to another. IMU can be used in locations with no GPS signals and also it is a cheaper and lighter device. Some works combine both GPS and IMU information for better data accuracy. Most smartphones, tablets and fitness tracking devices contain a low-cost IMU.